U.S. patent number 10,462,433 [Application Number 15/967,678] was granted by the patent office on 2019-10-29 for image sensor with big and small pixels and method of manufacture.
This patent grant is currently assigned to OmniVision Technologies, Inc.. The grantee listed for this patent is OmniVision Technologies, Inc.. Invention is credited to Gang Chen, Duli Mao, Dyson Hsin-Chih Tai.
![](/patent/grant/10462433/US10462433-20191029-D00000.png)
![](/patent/grant/10462433/US10462433-20191029-D00001.png)
![](/patent/grant/10462433/US10462433-20191029-D00002.png)
![](/patent/grant/10462433/US10462433-20191029-D00003.png)
![](/patent/grant/10462433/US10462433-20191029-D00004.png)
![](/patent/grant/10462433/US10462433-20191029-D00005.png)
![](/patent/grant/10462433/US10462433-20191029-D00006.png)
![](/patent/grant/10462433/US10462433-20191029-D00007.png)
United States Patent |
10,462,433 |
Chen , et al. |
October 29, 2019 |
Image sensor with big and small pixels and method of
manufacture
Abstract
An image sensor includes a substrate, a first set of sensor
pixels formed on the substrate, and a second set of sensor pixels
formed on the substrate. The sensor pixels of the first set are
arranged in rows and columns and are configured to detect light
within a first range of wavelengths (e.g., white light). The sensor
pixels of the second set are arranged in rows and columns and are
each configured to detect light within one of a set of ranges of
wavelengths (e.g., red, green, and blue). Each range of wavelengths
of the set of ranges of wavelengths is a subrange of said first
range of wavelengths, and each pixel of the second set of pixels is
smaller than each pixel of the first set of pixels.
Inventors: |
Chen; Gang (San Jose, CA),
Mao; Duli (Sunnyvale, CA), Tai; Dyson Hsin-Chih (San
Jose, CA) |
Applicant: |
Name |
City |
State |
Country |
Type |
OmniVision Technologies, Inc. |
Santa Clara |
CA |
US |
|
|
Assignee: |
OmniVision Technologies, Inc.
(Santa Clara, CA)
|
Family
ID: |
60808078 |
Appl.
No.: |
15/967,678 |
Filed: |
May 1, 2018 |
Prior Publication Data
|
|
|
|
Document
Identifier |
Publication Date |
|
US 20180249136 A1 |
Aug 30, 2018 |
|
Related U.S. Patent Documents
|
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
Issue Date |
|
|
15197464 |
Jun 29, 2016 |
9986213 |
|
|
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
H01L
27/14621 (20130101); H04N 5/3696 (20130101); H04N
9/045 (20130101); H04N 5/2253 (20130101); H04N
5/35563 (20130101); H01L 27/1462 (20130101); H04N
5/2257 (20130101); H01L 27/14605 (20130101); H04N
9/04555 (20180801); H04N 5/2252 (20130101); H01L
27/14627 (20130101); H01L 27/14645 (20130101); H04N
9/083 (20130101) |
Current International
Class: |
H04N
9/083 (20060101); H04N 9/04 (20060101); H04N
5/369 (20110101); H01L 27/146 (20060101); H04N
5/355 (20110101); H04N 5/225 (20060101) |
References Cited
[Referenced By]
U.S. Patent Documents
Primary Examiner: Peterson; Christopher K
Attorney, Agent or Firm: Henneman, Jr.; Larry E. Henneman
& Associates, PLC
Parent Case Text
CROSS-REFERENCE TO RELATED APPLICATIONS
This application is a divisional of co-pending U.S. patent
application Ser. No. 15/197,464, filed on Jun. 29, 2016 by the same
inventors, which is incorporated herein by reference in its
entirety.
Claims
We claim:
1. A method of manufacturing an image sensor, comprising: providing
a substrate; forming a first set of sensor pixels on said substrate
arranged in rows and columns; forming a second set of sensor pixels
on said substrate arranged in rows and columns, each pixel of said
second set of pixels being smaller than each pixel of said first
set of pixels; forming a set of transparent windows over said first
set of sensor pixels, said transparent windows being arranged in
rows and columns and each being configured to pass light within a
first range of wavelengths; and forming a set of filters over said
second set of sensor pixels, said filters being arranged in rows
and columns and each being configured to pass light within one of a
set of ranges of wavelengths, each range of wavelengths of said set
of ranges of wavelengths being a subrange of said first range of
wavelengths; wherein each pixel of said first set of pixels has a
center disposed between adjacent rows of said second set of pixels
and between adjacent columns of said second set of pixels; and
wherein said step of forming a second set of sensor pixels includes
forming each pixel of said second set of sensor pixels spaced apart
from every other pixel of said second set of sensor pixels by a
distance greater than a width of one of said sensor pixels of said
second set of sensor pixels.
2. The method of claim 1, wherein said set of ranges of wavelengths
includes a second range of wavelengths, a third range of
wavelengths, and a fourth range of wavelengths.
3. The method of claim 2, wherein: said second range of wavelengths
corresponds to a red portion of the visible light spectrum; said
third range of wavelengths corresponds to a green portion of the
visible light spectrum; and said fourth range of wavelengths
corresponds to a blue portion of the visible light spectrum.
4. The method of claim 3, wherein said step of forming a set of
filters over said second set of sensor pixels includes: forming a
first group of filters configured to pass only light within said
second range of wavelengths, said first group of filters comprising
25% of said second set of filters; forming a second group of
filters configured to pass only light within said third range of
wavelengths, said second group of filters comprising 50% of said
second set of filters; and forming a third group of filters
configured to pass only light within said fourth range of
wavelengths, said third group of filters comprising 25% of said
second set of filters.
5. The method of claim 4, wherein said step of forming a set of
filters includes: forming one half of said rows of said set filters
with filters of said first group and with filters of said second
group arranged in an alternating pattern; and forming the other
half of said rows of said set of filters with filters of said
second group and with filters of said third group arranged in an
alternating pattern.
6. The method of claim 5, wherein said step of forming a set of
filters includes: forming half of said columns of said set of
filters with filters of said first group and with filters of said
second group arranged in an alternating pattern; and forming the
other half of said columns of said set of filters with filters of
said second group and with filters of said third group arranged in
an alternating pattern.
7. The method of claim 1, wherein said step of forming a second set
of sensor pixels includes forming each pixel of said second set of
sensor pixels between each group of 4 adjacent pixels of said first
set of sensor pixels.
8. The method of claim 7, wherein said step of forming a set of
filters includes forming each filter of said set of filters between
each group of 4 adjacent transparent windows of said set of
transparent windows.
9. The method of claim 8, wherein said step of forming a second set
of sensor pixels includes forming each pixel of said second set of
sensor pixels between truncated corners of 4 adjacent pixels of
said first set of sensor pixels.
10. The method of claim 9, wherein said step of forming a set of
filters includes forming each filter of said set of filters between
truncated corners of 4 adjacent transparent windows of said set of
transparent windows.
11. The method of claim 1, wherein said step of forming a set of
filters includes forming each filter of said set of filters spaced
apart from every other filter of said set of filters by a distance
greater than a width of one of said filters.
12. The method of claim 1, wherein said steps of forming a first
set of sensor pixels and forming a second set of sensor pixels
include: forming one fewer rows of said sensor pixels of said
second set than said sensor pixels of said first set; and forming
one fewer columns of said sensor pixels of said second than said
sensor pixels of said first set.
13. The method of claim 12, wherein said steps of forming a set of
transparent windows and forming a set of filters include: forming
one fewer rows of said filters than said transparent windows; and
forming one fewer columns of said filters than said transparent
windows.
14. The method of claim 1, further comprising forming a set of
micro-lenses above said set of transparent windows and said set of
filters, said set of micro-lenses being configured to increase the
effective area of each pixel of said first set and said second set
of sensor pixels.
15. The method of claim 1, wherein said step of forming a second
set of sensor pixels includes forming said second set of sensor
pixels with a pitch between adjacent rows of said second set of
sensor pixels that is equal to a pitch between adjacent rows of
said first set of sensor pixels.
16. The method of claim 15, wherein said step of forming a second
set of sensor pixels includes forming said second set of sensor
pixels with a pitch between adjacent columns of said second set of
sensor pixels that is equal to a pitch between adjacent columns of
said first set of sensor pixels.
17. The method of claim 16, wherein said step of forming a set of
filters includes forming said set of filters with a pitch between
adjacent rows of said set of filters that is equal to a pitch
between adjacent rows of said set of transparent windows.
18. The method of claim 17, wherein said step of forming a set of
filters includes forming said set of filters with a pitch between
adjacent columns of said set of filters that is equal to a pitch
between adjacent rows of said set of transparent windows.
19. A method for manufacturing a camera module, comprising:
providing a housing defining an aperture; disposing a lens in said
aperture; and fixing an image sensor within said housing; and
wherein said image sensor is disposed in a focal plane of said
lens; said image sensor includes a plurality of image sensor pixels
including a first set arranged in rows and columns and a second set
arranged in rows and columns, each pixel of said first set
configured to detect light of the entire visible spectrum and each
pixel of said second set operative to detect light of one of three
distinct colors, each pixel of said first set being larger than
each pixel of said second set; each of said first set of pixels has
a center disposed between adjacent rows of said second set of
pixels and between adjacent columns of said second set of pixels;
and wherein said step of forming a second set of sensor pixels
includes forming each pixel of said second set of sensor pixels
spaced apart from every other pixel of said second set of sensor
pixels by a distance greater than a width of one of said sensor
pixels of said second set of sensor pixels.
Description
BACKGROUND OF THE INVENTION
Field of the Invention
This invention relates generally to digital imaging, and more
particularly to high dynamic range (HDR) pixel arrays.
Description of the Background Art
Color sensitive digital imagers are well known. Typically color
imagers are comprised of a plurality of image sensor pixels, each
being sensitive to light of one of three different colors. The
pixels are typically laid out in rows and columns, wherein each
square of four pixels includes one pixel sensitive to two of the
three colors and two pixels sensitive to the third color. Through
this arrangement, color images are created by interpolating missing
color information from neighboring pixels of that color. Color
images created in this way have deficiencies in quality, which
result from filtering out a large portion of the incident light and
filling in missing color information, such as poor light
sensitivity and low resolution.
One solution for increasing the quality of color images is the
utilization of a big-small pixel scheme for laying out pixels on an
imager sensor. In the big-small pixel scheme, some of the pixels
are larger than others. In prior art systems, both big and small
pixels are sensitive to one of three colors (e.g., R, G, B). The
big pixels are more sensitive to lower light intensities than the
small pixels. The big-small scheme facilitates the capture of
images with a higher dynamic range (HDR) and provides more detail
in images with widely varying local light intensities. In areas of
the image with high light intensity, the small pixels will be
sampled, because they are less sensitive to the light and will,
therefore, be less likely to have become saturated. In areas of the
image with lower light intensity, the big pixels will be sampled,
because they are more sensitive to the light. The resultant image
has a higher dynamic range, and shows more detail in both high and
low light areas. However, big-small schemes are not preferable for
low light applications, because a significant portion of the
impinging light is filtered out by the color filters which pass
only a small spectral band compared to the entire visible
spectrum.
Another solution for increasing the quality of color images is the
utilization of polychromatic (i.e. white) image sensing pixels in
addition to the color sensing pixels. These pixels are sensitive to
light across most of the visible spectrum, increasing the light
sensitivity of the white pixels as compared to the color filtered
pixels. Several white/color pixel layouts have been utilized,
including a big-small scheme, in which all the small pixels are
polychromatic (e.g., white) and all the big pixels are color
sensitive (e.g., R, G, B). This layout has been used, for example
to reduce color aliasing in a recorded image. Another layout
utilizes one each of the three color sensitive pixels and a
polychromatic pixel arranged in a square, which is repeated. Yet
another layout utilizes hexagonal pixels to create a layout with
twice as many polychromatic light sensing pixels as color sensing
pixels. Yet another layout utilizes a 4.times.4 square of image
sensing pixels wherein the color pixels are arranged along a
diagonal of the square (G-R-G-B) and the remaining pixels of the
square are polychromatic. These layouts each improve image
resolution and sensitivity, but there is still a desire for even
better image quality.
Yet another solution for increasing the quality of color images is
the fusion of a monochromatic and color version of the same image,
taken with separate image sensors. The monochromatic image, because
all the pixels are the same, has a resolution that is at least
three times higher than the color image. Using the monochromatic
image as the base, the two are used to generate a resulting color
image that has a higher resolution and lower noise than a
traditional color image. Drawbacks to this solution include the
increased costs and complexity due to the utilization of two
separate imagers. This solution also requires that the two images
be realigned digitally, which requires increased processing power
and sacrifices speed. Additionally, this solution sacrifices low
light sensitivity, because the two sensors have pixels of the same
size.
What is needed, therefore, is a single color imager that produces
images with an even greater sensitivity, resolution, and/or dynamic
range than the prior art.
SUMMARY
The present invention overcomes the problems associated with the
prior art by providing an image sensor including arrays of large
(black and white) and small (color) pixels. The invention
facilitates better image resolution, higher sensitivity, and
increased dynamic range.
An example image sensor includes a substrate, a first set of sensor
pixels and a second set of sensor pixels formed on the substrate.
Sensor pixels of the first set are arranged in rows and columns and
are each configured to detect light within a first range of
wavelengths. Sensor pixels of the second set are arranged in rows
and columns and are each configured to detect light within one of a
set of ranges of wavelengths. Each range of wavelengths of the set
of ranges of wavelengths is a subrange of the first range of
wavelengths. In addition, each pixel of the second set of pixels is
smaller than each pixel of the first set of pixels, and each pixel
of the first set of pixels has a center disposed between adjacent
rows of the second set of pixels and between adjacent columns of
the second set of pixels.
In the example embodiment, the set of ranges of wavelengths
includes a second range of wavelengths, a third range of
wavelengths, and a fourth range of wavelengths. In a particular
embodiment, the second range of wavelengths corresponds to a red
portion of the visible light spectrum, the third range of
wavelengths corresponds to a green portion of the visible light
spectrum, and the fourth range of wavelengths corresponds to a blue
portion of the visible light spectrum. The second set of sensor
pixels includes a first group of color pixels, a second group of
color pixels, and a third group of color pixels. The first group of
color pixels is configured to detect only light within the second
range of wavelengths, and the first group makes up 25% of the
second set of sensor pixels. The second group of color pixels is
configured to detect only light within the third range of
wavelengths, and the second group makes up 50% of the second set of
sensor pixels. The third group of color pixels is configured to
detect only light within the fourth range of wavelengths, and the
third group makes up 25% of the second set of sensor pixels.
In an example image sensor, each pixel of the first set of sensor
pixels includes a transparent window operative to pass all light
within the first range of wavelengths. Each pixel of the first
group of the second set of sensor pixels includes a light filter
operative to pass light within the second range of wavelengths.
Each pixel of the second group of the second set of sensor pixels
includes a light filter operative to pass light within the third
range of wavelengths, and each pixel of the third group of the
second set of sensor pixels includes a light filter operative to
pass light within the fourth range of wavelengths.
In an example pixel arrangement, one half of the rows of the second
set of sensor pixels include pixels of the first group and pixels
of the second group of the second set of sensor pixels arranged in
an alternating pattern. The other half of the rows of the second
set of sensor pixels include pixels from the second group and the
third group of the second set of sensor pixels arranged in an
alternating pattern. In addition, one half of the columns of the
second set of sensor pixels include pixels from the first group and
pixels from the second group of the second set of sensor pixels
arranged in an alternating pattern. The other half of the columns
of the second set of sensor pixels include pixels from the second
group and pixels from the third group of the second set of sensor
pixels arranged in an alternating pattern.
In an example embodiment, only one of the second set of sensor
pixels is disposed between each group of 4 adjacent pixels of the
first set of sensor pixels. In a more particular embodiment, each
of the second set of sensor pixels is disposed between truncated
corners of 4 adjacent pixels of the first set of sensor pixels.
Each of the sensor pixels of the first set is substantially
rectangular, with truncated corners, and each of the sensor pixels
of the second set is substantially rectangular and oriented at an
angle with respect to the pixel sensors of the first set. Side
edges of the sensor pixels of the second set are substantially
parallel to edges of the truncated corners of the sensor pixels of
the first set.
In an example embodiment, each of the second set of sensor pixels
is spaced apart from every other one of the second set of sensor
pixels by a distance greater than a width of one of the second set
of sensor pixels. The pitch between adjacent columns of the first
set of sensor pixels is the same as a pitch between adjacent
columns of the second set of sensor pixels, and the pitch between
adjacent rows of the first set of sensor pixels is the same as a
pitch between adjacent rows of the second set of sensor pixels. The
first set of sensor pixels includes one fewer rows than the second
set of sensor pixels, and the first set of sensor pixels includes
one fewer columns than the second set of sensor pixels.
In an example embodiment, the substrate is a silicon substrate, and
the image sensor includes a set of micro-lenses disposed above the
first set of sensor pixels and the second set of sensor pixels and
configured to increase the effective area of each pixel of the
first set and the second set of sensor pixels.
An example filter array for an image sensor is also disclosed. The
filter array includes a set of transparent windows and a set of
filters. The transparent windows are disposed above the image
sensor, arranged in rows and columns, and configured to pass light
within a first range of wavelengths. The filters are disposed above
the image sensor, arranged in rows and columns, and each configured
to pass light within one of a set of ranges of wavelengths. Each
range of wavelengths of the set of ranges of wavelengths is a
subrange of the first range of wavelengths, and each filter of the
set of filters is smaller than each transparent window of the set
of transparent windows. Each transparent window of the set of
transparent windows has a center disposed between adjacent rows of
the filters and between adjacent columns of the filters.
The set of ranges of wavelengths includes a second range of
wavelengths, a third range of wavelengths, and a fourth range of
wavelengths. The second range of wavelengths corresponds to a red
portion of the visible light spectrum, the third range of
wavelengths corresponds to a green portion of the visible light
spectrum, and the fourth range of wavelengths corresponds to a blue
portion of the visible light spectrum.
In the example filter array, the set of filters includes a first
group of color filters, a second group of color filters, and a
third group of color filters. The filters of the first group of
color filters are configured to pass only light within the second
range, and make up 25% of the set of filters. The filters of the
second group of color filters are configured to pass only light
within the third range, make up 50% of the set of filters. The
filters of the third group of color filters are configured to pass
only light within the fourth range, and make up 25% of the set of
filters.
In the example filter, one half of the rows of the set of filters
include filters from the first group and filters from the second
group of the set of filters arranged in an alternating pattern. The
other half of the rows of the set of filters include filters from
the second group and filters from the third group of the set of
filters arranged in an alternating pattern. Similarly, one half of
the columns of the set of filters include filters from the first
group and filters from the second group of the set of filters
arranged in an alternating pattern. The other half of the columns
of the set of filters include filters from the second group and
filters from the third group of the set of filters arranged in an
alternating pattern.
In an example arrangement, only one filter of the set of filters is
disposed between each group of 4 adjacent transparent windows of
the set of transparent windows, and each filter is spaced apart
from every other filter by a distance greater than a width of one
of the filters. Each filter of the set of filters is disposed
between truncated corners of 4 adjacent transparent windows of the
set of transparent windows. Each of the transparent windows is
substantially rectangular and has truncated corners. Each of the
filters is substantially rectangular and oriented at an angle with
respect to the transparent windows, such that side edges of the
filters are substantially parallel to edges of the truncated
corners of the transparent windows. The set of transparent windows
includes one fewer rows than the set of filters, and the set of
transparent windows includes one fewer columns than the set of
filters. A pitch between adjacent columns of the set of transparent
windows is the same as a pitch between adjacent columns of the set
of filters, and a pitch between adjacent rows of the set of
transparent windows is the same as a pitch between adjacent rows of
the set of filters.
Optionally, the filter array includes a set of micro-lenses
disposed above the set of transparent windows and above the set of
filters. The micro lenses are configured to increase the effective
area of the image sensor.
An example camera module is also disclosed. The example cameral
module includes a housing, a lens, and an image sensor. The housing
defines an aperture, and the lens is disposed in the aperture. The
image sensor is fixed within the housing and is disposed in a focal
plane of the lens. The image sensor includes a plurality of image
sensor pixels including a first set arranged in rows and columns
and a second set arranged in rows and columns. Each pixel of the
first set is configured to detect light of the entire visible
spectrum, and each pixel of the second set is operative to detect
light of one of three distinct colors. The pixels of the first set
are larger than each pixel of the second set, and each pixel of the
first set of pixels has a center disposed between adjacent rows of
the second set of pixels and between adjacent columns of the second
set of pixels.
An example method of manufacturing an image sensor is also
disclosed. The example method includes providing a substrate,
forming a first set of sensor pixels on the substrate arranged in
rows and columns, and forming a second set of sensor pixels on the
substrate arranged in rows and columns. Each pixel of the second
set of pixels is smaller than each pixel of the first set of
pixels.
The example method further includes forming a set of transparent
windows over the first set of sensor pixels. The transparent
windows are arranged in rows and columns and each window is
configured to pass light within a first range of wavelengths. The
example method further includes forming a set of filters over the
second set of sensor pixels. The filters are arranged in rows and
columns, and each filter configured to detect light within one of a
set of ranges of wavelengths, each range of wavelengths of the set
of ranges of wavelengths being a subrange of the first range of
wavelengths. Each pixel of the first set of pixels has a center
disposed between adjacent rows of the second set of pixels and
between adjacent columns of the second set of pixels.
The set of ranges of wavelengths includes a second range of
wavelengths, a third range of wavelengths, and a fourth range of
wavelengths. The second range of wavelengths corresponds to a red
portion of the visible light spectrum. The third range of
wavelengths corresponds to a green portion of the visible light
spectrum, and the fourth range of wavelengths corresponds to a blue
portion of the visible light spectrum.
In an example method, the step of forming a set of filters over the
second set of sensor pixels includes forming a first group of
filters, forming a second group of filters, and forming a third
group of filters. The first group of filters is configured to pass
only light within the second range of wavelengths and makes up 25%
of the second set of filters. The second group of filters is
configured to pass only light within the third range of wavelengths
and makes up 50% of the second set of filters. The third group of
filters is configured to pass only light within the fourth range of
wavelengths and makes up 25% of the second set of filters.
In a particular method, the step of forming a set of filters
includes forming one half of the rows of the set of filters with
filters of the first group and with filters of the second group
arranged in an alternating pattern. The step of forming a set of
filters further includes forming the other half of the rows of the
set of filters with filters of the second group and with filters of
the third group arranged in an alternating pattern. The step of
forming a set of filters further includes forming one half of the
columns of the set of filters with filters of the first group and
with filters of the second group arranged in an alternating
pattern, and forming the other half of the columns of the set of
filters with filters of the second group and with filters of the
third group arranged in an alternating pattern.
In an example method, the step of forming a second set of sensor
pixels includes forming each pixel of the second set of sensor
pixels between each group of 4 adjacent pixels of the first set of
sensor pixels. In a more particular method, the step of forming a
second set of sensor pixels includes forming each pixel of the
second set of sensor pixels between truncated corners of 4 adjacent
pixels of the first set of sensor pixels. In addition, the step of
forming a set of filters includes forming each filter of the set of
filters between each group of 4 adjacent transparent windows of the
set of transparent windows. In the more particular method, the step
of forming a set of filters includes forming each filter of the set
of filters between truncated corners of 4 adjacent transparent
windows of the set of transparent windows.
In an example method, the step of forming a second set of sensor
pixels includes forming each pixel of the second set of sensor
pixels spaced apart from every other pixel of the second set of
sensor pixels by a distance greater than a width of one of the
sensor pixels of the second set of sensor pixels. In addition, the
step of forming a set of filters includes forming each filter of
the set of filters spaced apart from every other filter of the set
of filters by a distance greater than a width of one of the
filters.
In an example method, the steps of forming a first set of sensor
pixels and forming a second set of sensor pixels include forming
one fewer rows of the sensor pixels of the second set than the
sensor pixels of the first set, and forming one fewer columns of
the sensor pixels of the second than the sensor pixels of the first
set. The steps of forming a set of transparent windows and forming
a set of filters include forming one fewer rows of the filters than
the transparent windows, and forming one fewer columns of the
filters than the transparent windows.
Optionally, the example method further includes forming a set of
micro-lenses above the set of transparent windows and the set of
filters. The set of micro-lenses is configured to increase the
effective area of each pixel of the first set and the second set of
sensor pixels.
In an example method, the step of forming a second set of sensor
pixels includes forming the second set of sensor pixels with a
pitch between adjacent rows of the second set of sensor pixels that
is equal to a pitch between adjacent rows of the first set of
sensor pixels. In addition, the step of forming a second set of
sensor pixels includes forming the second set of sensor pixels with
a pitch between adjacent columns of the second set of sensor pixels
that is equal to a pitch between adjacent columns of the first set
of sensor pixels. The step of forming a set of filters includes
forming the set of filters with a pitch between adjacent rows of
the set of filters that is equal to a pitch between adjacent rows
of the set of transparent windows and forming the set of filters
with a pitch between adjacent columns of the set of filters that is
equal to a pitch between adjacent rows of the set of transparent
windows.
BRIEF DESCRIPTION OF THE DRAWINGS
The present invention is described with reference to the following
drawings, wherein like reference numbers denote substantially
similar elements:
FIG. 1 is a perspective view of a camera module mounted on a
circuit board of a host device;
FIG. 2 is a block diagram of the camera module of FIG. 1;
FIG. 3 shows an example pixel layout for the image sensor of FIG.
1;
FIG. 4 is an exploded view of a portion of the image sensor array
of FIG. 2;
FIG. 5 is a diagram of the circuitry of the image sensor array of
FIG. 2;
FIG. 6 is a diagram of the circuitry of a pixel of the image sensor
array of FIG. 2; and
FIG. 7 is a flow chart summarizing an example method of
manufacturing an image sensor.
DETAILED DESCRIPTION
The present invention overcomes the problems associated with the
prior art, by providing an image sensor that produces images with
high resolution, high dynamic range, and good low light
sensitivity. In the following description, numerous specific
details are set forth (e.g., sensor pixel circuit design) in order
to provide a thorough understanding of the invention. Those skilled
in the art will recognize, however, that the invention may be
practiced apart from these specific details. In other instances,
details of well-known image sensor practices (e.g., image data
processing, semiconductor fabrication, etc.) and components have
been omitted, so as not to unnecessarily obscure the present
invention.
FIG. 1 is a perspective view of an image sensor 100 mounted on a
portion of a printed circuit board (PCB) 102 that represents a PCB
of a camera host device (e.g., automobile, manufacturing machine,
security camera, medical device, cell phone, etc.). Image sensor
100 communicates electronically with other components of the host
device, via a plurality of conductive traces 104. In the example
embodiment, image sensor 100 is depicted as being part of a camera
module 106 that further includes an optical assembly 108 and a
housing 110. As shown, housing 110 is mounted over image sensor
100, and optical assembly 108 is beneath an optical aperture formed
in housing 110. Those skilled in the art will recognize that the
particular designs and/or presence of PCB 102, traces 104, optical
assembly 108, and housing 110 will depend on the particular
application, and are not particularly germane to the present
invention. Therefore, PCB 102, traces 104, optical assembly 108,
and housing 110 are representational in character only.
FIG. 2 is a block diagram showing image sensor 100 in greater
detail to include an image sensor array 200, a row controller 202,
a sampling circuit 204, an image processor 206, and a control
circuit 208. Image sensor array 200 is an array of large
polychromatic and small color (i.e. sensitive to narrower bands of
the visible light spectrum) image sensor pixels, arranged in rows
and columns. The large polychromatic image sensor pixels are
sensitive to a majority of the visible light spectrum (e.g., white
light). The small color image sensor pixels are sensitive to
narrower bands (e.g., particular colors, red, green, blue, etc.) of
the visible light spectrum.
Row controller 202 provides row control signals to image sensor
array 200, via a row bus 210, one at a time. When a row is
selected, each pixel in that row outputs information indicative of
the intensity of light that has impinged on that pixel during a
shutter period to sampling circuit 204, via a data bus 212.
Sampling circuit 204 captures and transfers each row of image data
to image processor 206, via a data bus 214, for further processing.
In order to capture a complete image, every row of image sensor
array 200 is individually selected and read from.
Image processor 206 is operative to convert the digital data
acquired by sampling circuit 204 into readable image data, using
known image processing techniques in combination with processes
specifically adapted for the unique pixel scheme of the present
invention. Image processor 206 outputs the processed image data to
the camera host device, via data lines 215 and traces 104 (FIG. 1),
to the camera host device.
Control circuit 208 provides coordination and control of the
function of each component of image sensor 100. Control circuit 208
receives control signals (e.g., instructions to capture one or more
images) from the camera hosting device, via an interface bus 217
coupled to one or more of traces 104, and generates internal
control instructions based at least in part on the control
instructions from the host device. For example, control circuit 208
asserts control signals (e.g., a series of row addresses) onto
control lines 216, causing row controller 202 to process the rows
associated with each of the asserted row addresses. Control circuit
208 also asserts control signals onto control lines 218, causing
sampling circuit 204 to read sensor data being asserted on data
lines 212 by a selected row of image sensor array 200. After a row
of image data is sampled, control circuit 208 asserts additional
signals on control lines 218, causing sampling circuit 204 to
transfer the sampled image data to image processor 206. Control
circuit 208 also provides control signals, via a control bus 220,
to image processor 206, which cause image processor 206 to read the
image data from sample circuit 204 and to output the processed
image data to the camera host device at appropriate times.
FIG. 3 shows the layout of sensor pixels of a representative
portion of image sensor array 200, including a plurality of color
sensor pixels 300 and a plurality of white sensor pixels 302. Color
sensor pixels 300 (R, G, and B) and white sensor pixels 302 (W) are
arranged in rows and columns, with a row of color sensor pixels 300
between each row of white sensor pixels 302 and a column of color
pixels 300 between each column of white sensor pixels 302. Color
sensor pixels 300 are light sensing pixels, each with a particular
color filter disposed above it. In the example embodiment, color
sensor pixels 300 are one of red (R), green (G), or blue (B)
sensitive pixels, arranged in a Bayer array, in which 25% of the
pixels are red, 25% are blue, and 50% are green. The red and blue
light sensitive pixels alternate along every other diagonal of the
array, and the green light sensitive pixels are disposed along the
remaining diagonals. Every other row of sensor array 200 includes
either red and green pixels arranged in an alternating pattern, or
green and blue pixels arranged in an alternating pattern.
Similarly, every other column of sensor array 200 includes either
red and green pixels arranged in an alternating pattern, or green
and blue pixels arranged in an alternating pattern. In alternate
embodiments color sensor pixels 300 can be sensitive to a greater
or fewer number of colors, or to other different colors.
White sensor pixels 302 are larger than color sensor pixels 300
and, therefore, are more sensitive to light. Interspersing larger
white sensor pixels 302 with smaller color sensor pixels 300
improves the low-light capability and the dynamic range of image
sensor 100. Because they are more sensitive, white sensor pixels
302 provide better image quality in dark areas of an image or in
low light situations, while color sensor pixels 300 provide better
image quality in bright areas of the image or bright light
situations. In addition, combining the large white pixels 302 with
the smaller color pixels 300 reduces the prevalence of undesirable
artifacts, such as blooming, wherein saturated pixels bleed charge
to surrounding pixels resulting in a glare on the image.
The shapes and orientations of the sensor pixels are configured to
use as much surface of the sensor array as possible and maintain a
column and row structure. The column and row structure facilitates
reading rows of captured intensity data from the array via data
lines 212. Each white sensor pixel 302 is an irregular octagon
formed by truncating the corners of a square. Each color sensor
pixel 300 is square, but rotated by 45 degrees with respect to the
truncated square shape of white sensor pixels 302. Each color
sensor pixel 300 is disposed between the truncated corners of 4
adjacent white sensor pixels 302. Color sensor pixels 300 and white
sensor pixels 302 can also be arranged in other patterns that can
fill the sensor surface plane.
Sensor array 200 includes alternating rows and columns of color
sensor pixels 300 and white sensor pixels 302. The pitch of the
rows and columns of color sensors 300 is the same as the pitch
between the rows and columns of white sensor pixels 302. Each color
sensor pixel 300 has a center that is located between adjacent rows
and columns of white sensor pixels 302.
The relative sizes of white sensor pixels 302 and color sensor
pixels 300 can be altered to improve performance even more for
certain applications. For example, further increasing the size of
white sensor pixels 302 provides even greater sensitivity,
resulting in even better low light capability. However, when using
larger white sensor pixels 302 the resolution is diminished. The
manufacturer of an image sensor array can optimize the performance
of the array for particular applications by adjusting the relative
sizes of color sensor pixels 300 and white sensor pixels 302. For
example, a security camera might have to work well at night,
without producing high resolution images. In this case a
manufacturer would utilize white sensor pixels 302 that are
significantly larger than color sensor pixels 300. In the example
embodiment, the distance between adjacent color sensor pixels 300
is greater than the width of a color sensor pixel 300. In other
words, the white pixels 302 are at least twice as wide and/or at
least twice as long as the color sensor pixels 300.
FIG. 4 shows an exploded portion of image sensor array 200. Light
sensors 402 are disposed below a light filter array 404, which
includes a plurality of red filters 406, green filters 408, and
blue filters 410, and white filters 412, arranged as shown. Light
filter array 400 is formed above the electronic architecture of
light sensor pixels 402, and each filter corresponds to an
underlying light sensor 402. Each of filters 406, 408, and 410
transmit only light of certain wavelengths, which correspond to the
"color" of the light to be measured by light sensors below, and can
be made using various organic dyes embedded in a carrier medium.
White filters 412 include a window that is transparent to a
majority of the visible light spectrum, including the wavelengths
associated with red filters 406, green filters 408, and blue
filters 410. The window can include a transparent medium or even an
open space.
FIG. 5 is a schematic drawing of image sensor array 200. Each of
color sensor pixels 300 and white sensor pixels 302 is electrically
coupled to row control lines 210 and data lines 212. When row
controller 202 (FIG. 2) asserts a row select signal on one of row
control lines 210, each pixel in the associated row asserts a
signal corresponding to a detected light intensity on an associated
one of data lines 212. Sampling circuit 204 (FIG. 2) then latches
the row of data values being asserted on data lines 212 by the
selected row, thus capturing a line of image data. The process
continues for each row until data indicative of an entire image has
been sent to sampling circuit 204.
FIG. 5 is not drawn to scale. For example, significant space is
shown between neighboring pixels, to facilitate a clear view of row
control lines 210 and data lines 212. In actual embodiments, sensor
pixels 300 and 302 are as close together as possible, to maximize
the sensitive area of sensor array 200.
FIG. 6 is a circuit diagram of an individual image sensor pixel 600
(e.g., one of color sensing pixels 300 or white sensor pixels 302).
In the example embodiment, sensor pixel 600 is a four-transistor
(4T) pixel, including a photodiode 602, a transfer gate 604, a
floating diffusion region 606, a source follower 608, a select gate
610, and a reset gate 612. Row control lines 210 are shown to
include a select line (SEL) 614, a reset line (RST) 616, and a
transfer line (TX) 618.
Reset gate 612 includes a first terminal 620 electrically coupled
supply voltage Vdd, a second terminal 622 electrically coupled to
floating diffusion region 606, and a gate 624 electrically coupled
to reset line 616. When row controller 202 asserts a reset signal
(e.g. a digital high) on reset line 616, transistor 612 is
temporarily placed into a conducting state wherein floating
diffusion region 606 is coupled to voltage source Vdd. As a result,
the previous charge state of floating diffusion region 606 is
returned to a known reference charge state. Then, when row
controller 202 asserts a low voltage on reset line 616, reset
transistor 612 returns to a non-conducting state, whereby floating
diffusion region 606 is electrically isolated from voltage source
Vdd.
When light impinges on photodiode 602, negative charge carriers
build up. The amount of charge carriers that build up is based on
the intensity of the light impinging on photodiode 602 and the
length of time that the light is impinging.
Transfer gate 604 (a gate transistor) includes a first terminal
626, which is electrically coupled to photodiode 602, a second
terminal 628, which is electrically coupled to floating diffusion
region 606, and a gate terminal 630 which is electrically coupled
to transfer line 618. After a shutter period, row controller 202
asserts a signal on transfer line 618 and, therefore, on gate
terminal 630, which places transfer gate 604 in a conducting state,
allowing the accumulated charge to move from photodiode 602 to
floating diffusion region 606. The charge transfer pulls the
voltage on floating diffusion 606 down from its reset voltage (Vdd)
by an amount depending on the accumulated transferred charge.
Source follower 608 includes a gate terminal 632, which is
electrically coupled to floating diffusion 606, a second terminal
634, which is electrically coupled to a voltage source Vdd, and a
third terminal 636, which is electrically coupled select gate 610.
The voltage on third terminal 636 depends on the conductive state
of source follower 608, which depends on the voltage on gate
terminal 632. Therefore, the voltage on third terminal 636 will be
a known function of the voltage on floating diffusion region 606,
which depends on the charge transferred from the photodiode 602.
The voltage on the third terminal 636, therefore, represents the
intensity value generated by the pixel sensor 600.
Select gate 610 includes first terminal 638, which is electrically
coupled to third terminal 636, a second terminal 640, which is
electrically coupled to one of data control lines 212, and a gate
terminal 642, which is coupled to select line 614. When a signal is
asserted onto select line 614, that signal is conveyed to gate
terminal 642 select gate 610, placing select gate 610 into a
conductive state, thereby communicating the voltage on third
terminal 636 of source follower 608 to data line 212. This voltage
is latched by sampling circuit 204 (FIG. 2) and is indicative of
the intensity of light that impinged on photodiode 602 during the
shutter period.
FIG. 7 is a flow chart summarizing an example method 700 of
manufacturing an image sensor. In a first step 702, a substrate is
provided. Then, in a second step 704, a set of smaller sensor
pixels arranged in columns and rows is formed on the substrate.
Next, in a third step 706, a set of larger sensor pixels arranged
in columns and rows is formed on the substrate. Then, in a fourth
step 708, a set of transparent windows is formed over the larger
pixels, and, in a fifth step 710, a set of color filters is formed
over the smaller sensor pixels.
The description of particular embodiments of the present invention
is now complete. Many of the described features may be substituted,
altered or omitted without departing from the scope of the
invention. For example, alternate pixel shapes may be substituted
for the specific shapes shown. As another example, alternate sensor
circuitry (e.g., three-transistor sensor) can be substituted for
the four-transistor circuit described. These and other deviations
from the particular embodiments shown will be apparent to those
skilled in the art, particularly in view of the foregoing
disclosure.
* * * * *